IMPLEMENTATION OF THE ARIMA ALGORITHM FOR ENHANCING AND FORECASTING COMPANY REVENUE

Abstract

In an increasingly competitive business environment, accurate revenue forecasting is crucial for strategic decision-making. This study implements the ARIMA (AutoRegressive Integrated Moving Average) model to predict the monthly revenue of CV. Yusindo Mega Persada using historical data from January to December 2024. The ARIMA(1,1,1) model was selected based on stationarity tests and analysis of Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF) plots. Diagnostic tests confirmed that the model met key assumptions, including normality of residuals and absence of significant autocorrelation, ensuring reliable predictions. The forecasting results for January to March 2025 indicated a relatively stable revenue trend, with values ranging from IDR 483 million to IDR 489 million. Model accuracy was evaluated using the Mean Absolute Percentage Error (MAPE), which resulted in 9.41%, suggesting reasonable predictive performance. The findings demonstrate that ARIMA is capable of capturing trends and fluctuations in dynamic revenue data, providing actionable insights for management in financial planning, resource allocation, and risk mitigation. Despite the model’s effectiveness, external factors such as market fluctuations and seasonal events may influence actual revenue, indicating the need to combine quantitative forecasts with expert judgment. Overall, this study confirms that the ARIMA(1,1,1) model is a practical and reliable tool for revenue forecasting in a dynamic business environment.